kayuksel / forex-rl-challenge Goto Github PK

Hi nice work, thanks for sharing.
I am running example_cpu.py and getting lots of errors like the following

Process Process-21:
Traceback (most recent call last):
  File "/opt/conda/lib/python3.6/multiprocessing/process.py", line 258, in _bootstrap
    self.run()
  File "/opt/conda/lib/python3.6/multiprocessing/process.py", line 93, in run
    self._target(*self._args, **self._kwargs)
  File "example_cpu.py", line 72, in worker
    policy_d)
  File "example_cpu.py", line 32, in calculate_reward
    action = model.forward(state, policy) if torch.is_tensor(policy) else model.forward(state)
  File "example_cpu.py", line 144, in forward
    if torch.is_tensor(policy): return torch.matmul(state.double(), policy.double())
RuntimeError: size mismatch, m1: [1 x 518], m2: [515 x 4] at /pytorch/aten/src/TH/generic/THTensorMath.cpp:940

the code still runs but I am not sure what results I can get.
Why do we have m1 and m2 with different sizes?

Could you create some videos to teach us how it works and how to contribute for this?

I excitend to see this code since I started to learning Dqn and I hope I could test and contribute with something as soon as possible

Thanks for sharing, very interesting results. However, I don't quite get what the VAE plots mean. The return (c) is plotted against the first element of the latent vector. I don't see a strong correlation between the two, which indicats the movement is captured?

Hi,

Any description of the features used in this work and how to create them?

Cheers